International Journal of Modern Engineering Research (IJMER)

www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645

Nonlinear Control of UPFC in Power System for Damping Inter

Area Oscillations
1
M.Praveen Kumar, 2P B Chennaiah
1
EEE,M.Tech student, AITS, Rajampet, Kadapa , A.P, India,
2
EEE, Associate Professor, AITS, Rajampet,Kadapa, A.P, India,

Abstract: This paper introduces a new nonlinear control Therefore developing suitable control strategies is
of flexible ac transmission systems (FACTS) controllers for a requirement before FACTS can be confidently utilized in
the purpose of damping interarea oscillations in power the power system. Several authors have investigated
systems. FACTS controllers consist of series, shunt, or a utilizing FACTS, especially UPFCs to damp interarea
combination of series-shunt devices which are interfaced oscillations using a variety of control approaches [1]–[10].
with the bulk power system through injection buses. Interarea oscillations can occur in a system because of
Controlling the angle of these buses can effectively damp contingencies such as sudden load changes or faults. In [1]–
low frequency interarea oscillations in the system. The [5], oscillation damping is based on a linear control
proposed control method is based on finding an equivalent approach to the UPFC and power system, whereas other
reduced affine nonlinear system for the network from which authors consider nonlinear control systems theory and
the dominant machines are extracted based on dynamic Lyapunov Energy Functions [6]–[10]. Typically, nonlinear
coherency. It is shown that if properly selected, approaches are more effective for large perturbations or
measurements obtained from this subsystem of machines when the power system state strays significantly from the
are sufficient inputs to the FACTS controllers to stabilize initial operating point.
the power system. The effectiveness of the proposed method
on damping interarea oscillations is validated on the 68 The approach proposed in this paper provides
bus, 16 generator system of the New England/New York a general nonlinear method for using multiple FACTS
network. controllers in a power network for the purpose of damping
interarea oscillations. In this paper, it is shown that any
IndexTerms: Coherent groups, dominant machines, FACTS device capable of changing its interface bus
flexible ac transmission systems (FACTS), interarea angle(s) with the network can be used to mitigate power
oscillation, nonlinear control, phasor measurement unit system oscillations. Using this method, it will be shown that
(PMU), wide-area control. both shunt and series FACTS controllers can be used for
this purpose. The control method is based on finding a
I. INTRODUCTION reduced nonlinear affine state space system for the network
which can be controlled by feedback of selected
AS high voltage power electronics become less measurements of rotor frequencies. While frequency
expensive and have a wider-range of operation, flexible ac measurements (such as from FNET [11]) have made wide
transmission systems (FACTS) controllers will become area control of the power networks feasible, it is still not
more prevalent in the transmission system to control active reasonable to expect that the full set of frequency
power flow across congested corridors and ensure voltage measurements is available for controller use. Therefore, an
security. In addition, FACTS controllers can provide approach is proposed to use a reduced set of measurements
promising solutions to many of the stability problems that from a subset of machines in the system.
occur within the bulk power system. FACTS controllers can II. UPFC MODEL
be categorized into three major groups: shunt devices such
as the Static Synchronous Compensator (STATCOM), The UPFC is the most versatile FACTS device. It
series devices such as the static synchronous series consists of a combination of a shunt and series branches
compensator (SSSC) and series shunt devices such as the connected through the DC capacitor as shown in Fig. 1.
unified power flow controller (UPFC). In addition to Models for the STATCOM and SSSC can be easily
steady-state solutions such as power flow and voltage extracted from the UPFC model by considering the shunt
control, an added benefit of FACTS controllers deployed in and series converters individually. The series connected
the transmission system is that they can also effectively inverter injects a voltage with controllable magnitude and
Control active power oscillations that can damage phase angle in series with the transmission line, therefore
generators, increase line losses, and increase wear and tear providing active and reactive power to the transmission
on network components. line. The shunt-connected inverter

www.ijmer.com 4266 | Page

 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645

Fig.1. Unified power flow controller diagram.

Provides the active power drawn by the series branch plus

the losses and can independently provide reactive
compensation to the system. The UPFC model is given by
[12] as shown in (1)–(5) at the bottom of the next page,
where the parameters are as shown in Fig. 1. The currents
and are the components of the shunt
current. The currents and are the
components of the series current. The voltages
and are the sending end and receiving end
voltage magnitudes and angles, respectively. The UPFC is
controlled by varying the phase angles ( ) and (
Fig. 2. Equivalent power system from the controller
) magnitudes of the converter shunt and series viewpoint.
output voltages ( ), respectively. Using the load impedance model, the only points of current
The power balance equations at bus 1 are given by injection into the network are the generator internal buses
and the UPFC sending and receiving end buses. Using
Kron reduction, the transmission network can be reduced
to an admittance matrix of size ( ) whereN
………………(1) is the number of generator buses and is the number FACTS
current injections in the system. Fig. 2 illustrates the
reduced system showing the points of current injection.
……………. (2) Each UPFC has two current injections and , at the
sending and receiving ends, respectively; a STATCOM
and at bus 2 and SSSC have only one current. The generator current
injections are given by .
The classical model for the reduced network
……(3) including the UPFCs.
Mechanical power, inertia constant and angular speed,
respectively, of machine; and is synchronous speed. The
………(4) summation represents the active power injected at each
current injection point, including FACTS buses.

IV. CONTROLLER DESIGN

III. SYSTEM MODEL

For control development purposes, several The controller design consists of three stages.
initial assumptions are made. The first assumption is that A. Stage I
the system loads are modeled as constant impedance loads The objective of the first design stage is to find
the desired changes in mechanical powers required to
and can therefore be absorbed into the bus admittance
matrix. Second, the generators are modeled as the classical stabilize the system. To obtain the amount of mechanical
―transient reactance behind constant voltage‖ model. Note power required, it is initially assumed that the mechanical
that these assumptions are for control development only— powers ,are inputs into the system model. Note that
the proposed control is validated with the full nonlinear 10- this is only for controller development; in the final control,
th order power system model given in the Appendix [13]. it is not required that the generator mechanical powers
In addition, the proposed control is developed for the actually vary.
UPFC; control development for the STATCOM and SSSC Under this assumption, the system models are
follows a similar procedure and is therefore not explicitly ……………………… (5)
detailed.

www.ijmer.com 4267 | Page

 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645
where as shown in (13)–(15) at the bottom of the page and affine nonlinear system, the output 𝑦(𝑡) tracks a prescribed
. reference value 𝜔 𝑡 in terms of a given performance:
Since it is only required that the system frequencies return
to steady-state rapidly, a subset of (12) is 𝑥 = 𝑓(𝑥 𝑡 ) + 𝑔 𝑥 𝑡 𝑢(𝑡) ………..(14)
…………………….(6)
where and 𝑦𝑖 𝑡 = 𝑕𝑖 (𝑥 𝑡 ) i=1,….,m ………….(15)

Where 𝑚 is the number of outputs equal to the number of

inputs in 𝑢(𝑡) . The receding horizon performance index is
given by
where ..(7)
1
⋯ 0 1 𝑇
𝐽 = 2 0 (𝑦 (𝑡 + 𝜏) − 𝜔(𝑡 + 𝜏))𝑑𝜏 × (𝑦 𝑡 + 𝜏 − 𝜔(𝑡 +
𝑀1
𝑔 = ⋮ ⋱ ⋮ …………………….(8) 𝜏))𝑑𝜏 ………….(16)
1
0 ⋯ 𝑀1 whereT is the predictive period. The actual control
𝑢 = [𝑃𝑚1 𝑃𝑚2 … 𝑃𝑚𝑛 ] input 𝑢(𝑡) is given by the initial value of the optimal
Letting x1s ,x2s , and us denote the steady-state values control input 𝑢(𝑡 + 𝜏) for 0 ≤ 𝜏 ≤T and 𝑢(𝑡 + 𝜏) when
ofx1 ,x2 and u , respectively, then the error in generator 𝜏 =0.
rotor frequencies becomes The optimal predictive control law is given by
е = x2-x2s…………..(9)
𝑝−1 𝜌
and 𝑢 𝑡 = −(𝐿𝑔 𝑙𝑓 𝑕(𝑥))−1 (𝐾𝑀𝜌 + 𝐿𝑓 𝑕 𝑥 − 𝜔 𝜌
(𝑡)
ė = f(x)-f(x1s)-gus+gud……………(10) ……..(17)
Equation (21) can be stabilized with input 𝑢𝑑so that
𝑢𝑑 = 𝑔−1 [-f(𝑥1 )+f(𝑥1𝑠 )-𝑔𝑢𝑠 +Ke]………..(11) where𝜌 is the relative degree for the system outputs
(assuming that all outputs have the same relative degree)
where K is a positive definite matrix and and 𝐿 is the Lie derivative defined by
𝑒 . = −𝐾𝑒…………..(12)
B. Stage II 𝜕𝑣
𝐿𝜇 𝑣 = 𝜕𝑥 𝜇 ……..(18)
In Stage I, the required changes in the generator’s
mechanical powers were found that stablize the system. In
Stage II, these changes are translated into control signals to The matrix 𝐾 is given by
the FACTS controllers. As noted previously, the generator
mechanical powers do not actually change as a
consequence of the proposed control. Therefore, using the
desired active power changes, a new control signal is
introduced
⊿𝑢 = 𝑢𝑑𝑒𝑠𝑖𝑟𝑒𝑑 − 𝑢𝑎𝑐𝑡𝑢𝑎𝑙 ……….(13)
Where 𝑢𝑑𝑒𝑠𝑖𝑟𝑒𝑑 and 𝑢𝑎𝑐𝑡𝑢𝑎𝑙 are the desired and ………….(19)
actual values for the generator mechanical powers. This −1 𝑇
The matrix K is the first rows of the matrix 𝜓𝑟𝑟 𝜓𝜌𝑟 where
mismatch is translated into the desired changes in the
FACTS’ bus voltage angles, as shown in (25) at the bottom 𝜓(𝜌+1,𝜌+1) ⋯ 𝜓(𝜌+1,𝜌+𝑟+1)
of the page, where 𝜓𝑟𝑟 = ………(20)
𝜓(𝜌,𝜌+1) ⋯ 𝜓(𝜌+𝑟+1,𝜌+𝑟 +1)
𝐿 = [𝑙1,….………𝑙𝑁 ]𝑇
∆= [∆𝛿1 , … . . ∆𝛿𝑛 ]𝑇 𝜓(1,𝜌+1) ⋯ 𝜓(1,𝜌+𝑟+1)
The nonlinear system is solved numerically for ∆. 𝜓𝜌𝑟 = ……….(21)
𝜓(𝜌,𝜌+1) ⋯ 𝜓(𝜌,𝜌+𝑟+1)
Note that if 𝑁 ≠ 𝑛, then the system of equations is not Where
square and an exact solution is not possible. In this case,
the equations are solved to find the best fit to∆ which 𝑇 𝑖+𝑗 −1
minimizes the error. These values are then used to 𝜓𝑖,𝑗 = , i, j=1,….,𝜌 + 𝑟 + 1 ……..(22)
𝑖−1 ! 𝑗 −1 ! 𝑖 +𝑗 −1 !
calculate the desired current injections 𝑖 ∗𝑑1 ,𝑖 ∗𝑞1 ,𝑖 ∗𝑑2
,𝑖 ∗𝑞2 from the power balance (1)–(4).
And
C. Stage III 𝑇 = 𝑑𝑖𝑎𝑔 𝑇, … . . , 𝑇 𝜖 𝑅𝑚 ×𝑚
In Stage III, the desired current injections are Returning to (1)–(5), the relative degree for all of the
translated into actual control values for the FACTS outputs is 𝜌 = 1 and assuming the control order to be
controllers. As before, this approach is developed for the 𝑟 = 0, then the control law for the UPFC becomes
UPFC only, noting that similar approaches can be −3𝐿1 ∗
𝑅1 𝐿1
𝑢1 = 𝑖 − 𝑖𝑑1 + 𝑖 − 𝑖 +
developed for the SSSC and STATCOM. To find the 𝜔𝑠 𝑣𝑑𝑐 𝑇 𝑑 1 𝑉𝑑𝑐 𝑑1 𝑉𝑑𝑐 𝑞1
actual control inputs, a predictive control based on [14] is 𝑉1 cos 𝜃1
+
𝐿1 𝑑 ∗
𝑖𝑑1 ……….(23)
used. The basic methodology of predictive control is to 𝑉 𝑑𝑐 𝜔 𝑠 𝑉 𝑑𝑐 𝑑𝑡
design an asymptotically stable controller such that in an

www.ijmer.com 4268 | Page

 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645
−3𝐿1 ∗ 𝑅1 𝐿1 𝑉1 sin 𝜃1
𝑢2 = 𝑖 − 𝑖𝑞1 + 𝑖 − 𝑖 + +
𝜔 𝑠 𝑣𝑑𝑐 𝑇 𝑞1 𝑉 𝑑𝑐 𝑞1 𝑉 𝑑𝑐 𝑑1 𝑉 𝑑𝑐
𝐿1 𝑑 ∗
𝑖 …….(24)
𝜔 𝑠 𝑉 𝑑𝑐 𝑑𝑡 𝑞1

−3𝐿1 ∗ 𝑅 𝐿1 𝑉2 cos 𝜃2
𝑢3 = 𝑖 − 𝑖𝑑2 + 1 𝑖𝑑2 − 𝑖 + −
𝜔 𝑠 𝑣𝑑𝑐 𝑇 𝑑 2 𝑉 𝑑𝑐 𝑉 𝑑𝑐 𝑞2 𝑉 𝑑𝑐
𝑉 cos 𝜃1 𝐿2 𝑑 ∗
+ 1 𝑖 ……..(25)
𝑉 𝜔 𝑉 𝑑𝑡 𝑑2
𝑑𝑐 𝑠 𝑑𝑐

−3𝐿2 𝑅 𝐿 𝑉2 sin 𝜃2
𝑢4 = 𝜔 ∗
𝑖𝑞2 − 𝑖𝑞2 + 𝑉 2 𝑖𝑞2 − 𝑉 2 𝑖𝑑2 + − Fig. 3. Three stage control process.
𝑠 𝑣𝑑𝑐 𝑇 𝑑𝑐 𝑑𝑐 𝑉 𝑑𝑐
𝑉1 sin 𝜃1 𝐿1 𝑑 ∗
+𝜔 𝑖 𝑞1 ……..(26) 𝑢2
𝑉 𝑑𝑐 𝑠 𝑉 𝑑𝑐 𝑑𝑡 𝛼1 = tan−1 ………(28)
𝑢1

These inputs are then translated into the control inputs for 𝑘2 = 𝑢32 + 𝑢42 ……(29)
−1 𝑢 4
the UPFC 𝛼2 = tan ………….(30)
𝑢3
𝑘1 = 𝑢12 + 𝑢22 …………..(27)
The three stage control process and outcomes of each
stage are summarized in Fig.3

machines be ordered from 1 to Q and the rest of the

machines be numbered from Q+1 to N, then the changes in
the non-dominant machines can be approximated using a
zero-th order model by

𝑋𝑄+1,𝑄+1 ⋯ 𝑋𝑄+1,𝑁
∆𝛿𝑄+1
⋮ ⋱ ⋮ ⋮ =
𝑋𝑁,𝑄+1 ⋯ 𝑋𝑁,𝑁
∆𝛿𝑁
𝑄 𝑁+𝑛
𝑘=1 𝑋𝑄+1,𝑘 ∆𝛿𝑘 − 𝑘=𝑁+1 𝑋𝑄+1,𝑘 ∆𝛿𝑘
⋮ ….(31)
𝑄 𝑁+𝑛
𝑘=1 𝑋𝑁,𝑘 ∆𝛿𝑘 − 𝑘=𝑁+1 𝑋𝑁,𝑘 ∆𝛿𝑘
Where
Fig. 4. The 68 bus, 16 generator test system
𝜇 𝑖𝑗
𝑥𝑖,𝑗 = ……………(32)
𝜇 𝑖𝑖
V. SELECTIVEFEEDBACKMEASUREMENTS And
BASEDON DOMINANT MACHINES 𝜇𝑖𝑗 = −𝐸𝑖 𝐸𝑗 𝑌𝑖𝑗 sin(𝛿𝑖 − 𝛿𝑗 − 𝜑𝑖𝑗 ) i≠j …..(33)
The control method proposed in the previous 𝑁+𝑛
𝜇𝑖𝑖 = − 𝑘≠𝑖 𝐸𝑖 𝐸𝑗 𝑌𝑖𝑗 sin(𝛿𝑖 − 𝛿𝑗 − 𝜑𝑖𝑗 ) i=j ……….(34)
section requires generator rotor frequencies to be
implemented. Although with recent advances in wide area
frequency measurement (FNET) it may be possible to Note than when only the dominant machines are
provide synchronized global measurements, it is still not selected for the control action, only the rows corresponding
feasible to assume that all generator rotor frequencies are to the dominant machines will be used thereby reducing
simultaneously available. However, it is reasonable to the order of the system. This is advantageous since the
assume that a subset of the measurements are available for pseudo-inverse required to solve the set of equations is
feedback and the remainder of the states can be estimated more nearly square providing better convergence.
based on the available measurements. The most probable
machines to obtain measurements from are those machines VI. EXAMPLE AND RESULTS
which dominate coherent groups.
There are numerous methods for calculating Although the control has been developed
coherent groups in the literature [15]–[18]. In [18], the using the classical generator model, the control approach
coherency identification method is based on modal will be validated using the full 10th order model which
analysis and Gaussian elimination with full pivoting on the includes an exciter/AVR, turbine, and governor dynamics.
selected eigenvectors of the system to find the reference The model is given in the Appendix. The proposed control
generators and their group members. is validated on the 68 bus, 16 generator New England/New
The selected eigenvectors are chosen based on the York test system shown in Fig. 4. The coherent groupings
lowest oscillatory modes of the system. Once the dominant corresponding to the five slowest modes are indicated by
machines are found, a reduced order system is computed the dashed lines in Fig. 4. The network data and coherent
which captures the ―slow‖ dynamics of the original system. groupings are the same as in [19]. The transmission tie
In this process, the remaining unmeasured states of the lines are shown with bold lines. The reference generators
system can be estimated based on the states which are for the five areas are G5, G13, G14, G15, and G16.
measured via singular perturbation [13]. Let the dominant

www.ijmer.com 4269 | Page

 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645
TABLE I FACTS PARAMETERS that by definition, the shunt active power absorbed is
positive, thus during steady-state the STATCOM will
absorb active power and the figures indicate a positive
value. The shunt converter injects active power into the
system during the fault. Similar behavior is displayed by
Choosing the appropriate number of FACTS the STATCOMs as shown in Fig. 8. Fig. 9 shows the dc
controllers in the network is based on the number of link capacitor voltages. The UPFC and one of the
coherent areas. As a rule of thumb, it is best to match the STATCOMs experience a drop of approximately 5%
number of current injections with the number of modes. whereas the second STATCOM experiences a slight
For example, five current injections can be used to control increase in voltage. This is reasonable, since to damp
the interarea oscillations between five areas. In the 68 bus oscillations, it may be necessary to inject active power in
example, four current injections are used: one UPFC (two some areas and absorb active power in other areas.
injections) and two STATCOMS (one injection each). In
this paper, the UPFC is placed on line 1–2 with the shunt VII. CONCLUSIONS
converter on bus 2 and the STATCOMS have been placed A three stage nonlinear control scheme
on buses 47 and 49. The placements of the FACTS has been proposed for damping interarea oscillations using
controllers were heuristically chosen to be at buses at the multiple FACTS controllers. Any FACTS device that can
edge of the areas as might occur in practice. Several control its interface bu angle(s) with the power network
researchers have addressed the problem of optimal can utilize this control approach. The method uses the
placement of FACTS controllers. In [20], the authors generators’ frequencies as the feedback data for the
utilize modal sensitivity to determine placement of TCSCs. control. Using measurements from the dominant generators
Eigenvalue shift is used as a placement strategy for SVCs and estimating the rest of the states based on equivalent
in [21]. [22] focuses on the determination of the best bus reduced systems is shown to considerably reduce the
placement for SVCs to damp interarea oscillations. number of needed global measurements for control. Based
Another recent work addresses the use of modal on the simulation results, the proposed method shows
controllability indices specifically for FACTS placement promising results for wide-area control of power systems.
for oscillation damping [23]. There are several issues which need to be considered
The parameters of the FACTS controllers are however.
given in Table I. The per unit approach is the same as in There is a considerable computational burden for
[24] on a 100 MW, 100 kV base.\ the controller which requires fast processors for real-time
In the simulations, a solid three-phase fault is performance. However, good coherent groupings will
applied to bus 30 at 0.2 seconds and cleared at 0.3 seconds. lower the computation time by improving the estimation
The dynamic responses to this fault are shown for the process. Future work will also consider the effect of time
following cases: delays and communication noise in the measured states on
Case 1) proposed control, all measurements available; the control effectiveness. Sensitivity of the proposed
Case 2) proposed control, only dominant machine method to system uncertainties and topology changes will
measurements available; also be studied.
Case 3) linear control (taken from [25]).
Note that in Case 2), the estimation approach
proposed in Section V is used to obtain approximations to
the non-measured states.
Fig.5shows a subset of the generator speeds with
no FACTS controllers in the system compared to Case 1).
Not all responses are shown for the sake of brevity. The
selected generators are taken from four of the five coherent
areas (generator 15 is by itself in an area and is not shown).
Note that the generators go unstable as a result of the fault,
but the proposed control is able to stablize the system and
rapidly mitigate the oscillations.
Fig. 6 shows the active power injections of the
UPFC. The series injection is shown in the top figure and
the shunt injection is shown in the bottom figure. In this
figure, Case 2 (bold) is compared to Case 3 (thin). These
series active power injection for the proposed control is
very modest; therefore the rating ofthe series transformer
and converter do not need to be overly large. The shunt Fig. 5. Generator speeds for no FACTS controllers (bold)
active power is related to the series active power and Case 1 (thin).
𝑉2
−𝑃𝑠𝑕𝑢𝑛𝑡 = 𝑃𝑠𝑒𝑟𝑖𝑒𝑠 − 𝑅𝑑𝑐 ………..(35)
𝑑𝑐
Therefore 𝑃𝑠𝑕𝑢𝑛𝑡 will be opposite in polarity
to 𝑃𝑠𝑒𝑟𝑖𝑒𝑠 and will differ in magnitude by the losses in the
converter. Furthermore, during transients the dc link
capacitor will charge or discharge active power. Also note
www.ijmer.com 4270 | Page
 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645
[2] B. C. Pal, ―Robust damping of interarea oscillations
with unified power flowcontroller,‖ Proc. Inst. Elect.
Eng., Gen. Transm. Distrib., vol. 149, no. 6, pp. 733–
738, Nov. 2002.
[3] B. Chaudhuri, B. C. Pal, A. Zolotas, I. Jaimoukha,
and T. Green, ―Mixed-sensitivity approach to ��
control of power system oscillations employing
multiple FACTS devices,‖ IEEE Trans. Power Syst.,
vol. 18, no. 3, pp. 1149–1156, Aug. 2004.
[4] M. M. Farsangi, Y. H. Song, and K. Y. Lee, ―Choice
of FACTS device control inputs for damping
interarea oscillations,‖ IEEE Trans. Power Syst., vol.
19, no. 2, pp. 1135–1143, May 2004.
[5] N. Tambey and M. L. Kothari, ―Damping of power
system oscillations with unified power flow
controller (UPFC),‖ Proc. Inst. Elect. Eng., Gen.,
Transm. Distrib., vol. 150, pp. 129–140, Mar. 2003.
[6] M. Ghandhari, G. Andersson, and I. A. Hiskens,
―Control Lyapunov functions for controllable series
devices,‖ IEEE Trans. Power Syst., vol. 16, no. 4, pp.
Fig. 6. UPFC injected active power: Series (top) and shunt
689–694, Nov. 2001.
(bottom); Case 2 (bold) and Case 3 (dashed).
[7] S. Robak, M. Januszewski, and D. D.
Rasolomampionona, ―Power system stability
enhancement using PSS and Lyapunov-based
controllers: A comparative study,‖ in Proc. IEEE
Power Tech Conf., Bologna, Italy, 2003, vol. 3, p. 6.
[8] C.-C. Chu and H.-C. Tsai, ―Application of
Lyapunov-based adaptive neural network UPFC
damping controllers for transient stability
enhancement,‖ presented at the IEEE Power Energy
Soc. General Meet., Pittsburgh, PA, Jul. 20–24, 2008.
[9] A. Bidadfar, M. Abedi, M. Karari, and C.-C. Chu,
―Power swings damping improvement by control of
UPFC and SMES based on direct Lyapunov method
application,‖ presented at the IEEE Power and
Energy Soc. General Meet., Pittsburgh, PA, Jul. 20–
24, 2008.
[10] M. Januszewski, J. Machowski, and J. W. Bialek,
Fig. 7. STATCOM active power injection: Case 2 (bold),
―Application of th direct Lyapunov method to
Case 3 (dashed).
improve damping of power swings by control of
UPFC,‖ Proc. Inst. Elect. Eng., Gen., Transm.
Distrib., vol. 151, pp. 252–260, Mar. 2004.
[11] Z. Zhong, C. Xu, B. J. Billian, L. Zhang, S. J. Tsai,
R. W. Conners, V. A. Centeno, A. G. Phadke, and Y.
Liu, ―Power system frequency monitoring
network (FNET) implementation,‖ IEEE Trans.
Power Syst., vol. 20, no. 4, pp. 1914–1921, Nov.
2005.
[12] L. Dong, M. L. Crow, Z. Yang, and S. Atcitty, ―A
reconfigurable FACTS system for university
laboratories,‖ IEEE Trans. Power Syst., vol. 19, no.
1, pp. 120–128, Feb. 2004.
[13] P. Sauer and M. A. Pai, Power System Dynamics and
Fig. 8. FACTS Vdc: Case 2 (bold) and Case 3 (dashed). Stability. Upper Saddle River, NJ: Prentice-Hall,
1998.
[14] W.-H. Chen, D. J. Balance, and P. J. Gawthrop,
REFERENCES ―Optimal control of nonlinear systems:Apredictive
[1] H. Wang, ―A unified model for the analysis of control approach,‖ Automatica, vol. 39, pp. 633–641,
FACTS devices in damping power system 2003.
oscillations—Part III: Unified power flow [15] J. H. Chow, R. Galarza, P. Accari, and W. W. Price,
controller,‖ IEEE Trans. Power Del., vol. 15, no. 3, ―Inertial and slow coherency aggregation algorithms
pp. 978–983, Jul. 2000. for power system dynamic model reduction,‖ IEEE

www.ijmer.com 4271 | Page

 International Journal of Modern Engineering Research (IJMER)
www.ijmer.com Vol.2, Issue.6, Nov-Dec. 2012 pp-4266-4272 ISSN: 2249-6645
Trans. Power Syst., vol. 10, no. 2, pp. 680–685, May
1995.
[16] S. Geeves, ―A modal-coherency technique for
deriving dynamic equivalents,‖ IEEE Trans Power
Syst., vol. 3, no. 1, pp. 44–51, Feb. 1998.
[17] M. L. Ourari, L.-A. Dessaint, and V.-Q. Do,
―Dynamic equivalent modeling of large power M Praveen Kumar received B.Tech Degree in
Electrical & Electronics Engg. from JNT University, Hyderabad,
systems using structure preservation technique,‖ India. Presently he is with Annamacharya Institute of
IEEE Trans. Power Syst., vol. 21, no. 3, pp. 1284– Technology & Sciences, Rajampet, A.P,India, in Dept. of EEE
1295, Aug. 2006. and pursuing his M.Tech. His research interests includes power
[18] , J. H. Chow, Ed., Time-Scale Modeling of Dynamic systems, FACTS,
Networks With Applications to Power Systems.
Berlin, Germany: Springer-Verlag, 1982.
[19] G. Rogers, Power System Oscillations. Norwell, MA:
Kluwer, 2000.
[20] E. V. Larsen, J. J. Sanchez-Gasca, and J. H. Chow,
―Concepts for design of FACTS controllers to damp P.B Chennaiah received B.Tech Degree in
Electrical & Electronics Engg, from JNT University, Anantapur,
power swings,‖ IEEE Trans. Power Syst., vol. 10, no. India. M..Tech degree in electrical engineering from JNT
2, pp. 948–956, May 1995. University,Hyderabad. Presently he is with Annamacharya
[21] P. Pourbeik and M. J. Gibbard, ―Simultaneous Institute of Technology & Sciences, Rajampet, A.P,India, in
coordination of power system stabilizers and FACTS Dept. of EEE as an Associate professor. His research includes
device stabilizers in a multimachine power system optimization techniques in power systems, FACTS,
for enhancing dynamic performance,‖ IEEE Trans.
Power Syst., vol. 13, no. 2, pp. 473–479, May 1998.
[22] N. Martins and L. T. G. Lima, ―Determination of
suitable locations for power system stabilizers and
static VAR compensators for damping
electromechanical oscillations in large scale power
systes,‖ IEEE Trans. Power Syst., vol. 5, no. 4, pp.
1455–1469, Nov. 1990.
[23] B. K. Kumar, S. Singh, and S. Srivastava,
―Placement of FACTS controllers using modal
controllability indices to damp out power system
oscillations,‖ Proc. Inst. Eng. Technol., Gen.,
Transm., Distrib., vol. 1, no. 2, pp. 252–260, Mar.
2007.
[24] C. Schauder and H. Mehta, ―Vector analysis and
control of advanced static VAR compensators,‖ Proc.
Inst. Elect. Eng., C, vol. 140, no. 4, 1993.
[25] M. Zarghami and M. L. Crow, ―Discussion of
effective control of interarea oscillations by UPFCs,‖
presented at the 39th North American Power Symp.,
Las Cruces, NM, Oct. 2007.

www.ijmer.com 4272 | Page